Deep Learning Techniques for Voice Disorder Detection: A Transfer Learning and Data Augmentation Approach

Recognition of emotional states using frequency effective connectivity maps through transfer learning approach from electroencephalogram signals

Plant disease classification and prevention of spreading of the disease at earlier stages based on visual leaves symptoms and Pest recognition through deep learning-based image classification is in the forefront of research. To perform the investigation on Plant and pest classification, Transfer Learning (TL) approach is used on EfficientNet-V2. TL requires limited labelled data and shorter training time. However, the limitation of TL is the pre-trained model network’s topology is static and the knowledge acquired is detrimentally overwriting the old parameters. EfficientNet-V2 is a Convolutional Neural Network (CNN) model with significant high speed learning rates across variable sized datasets. The model employs a form of progressive learning mechanism which expands the network topology gradually over the course of training process improving the model’s learning capacity. This provides a better interpretability of the model’s understanding on the test domains. With these insights, our work investigates the effectiveness of EfficienetV2 model trained on a class imbalanced dataset for plant disease classification and pest recognition by means of combining TL and progressive learning approach. This Progressive Learning for TL (PL-TL) is used in our work consisting of 38 classes of PlantVillage dataset of crops and fruit species, 5 classes of cassava leaf diseases and another dataset with around 102 classes of crop pest images downloaded from popular dataset platforms, though it is not a benchmark dataset. To test the predictability rate of the model in classifying leaf diseases with similar visual symptoms, Mix-up data augmentation technique is used at the ratio of 1:4 on corn and tomato classes which has high probability of misinterpretation of disease classes. Also, the paper compares the TL approach performed on the above mentioned three types of data set using well established CNN based Inceptionv3, and Vision Transformer a non-CNN model. It clearly depicts that EfficientNetV2 has an outstanding performance of 99.5%, 97.5%, 80.1% on Cassava, PlantVillage and IP102 datasets respectively at a faster rate irrespective of the data size and class distribution as compared to Inception-V3 and ViT models.The performance metrics in terms of accuracy, precision, f1-score is also studied.

Manual inspection of fruit surface defects is time consuming, involves labour cost, prone to human error and possess inconsistent classification standards in fruit sorting application. To solve this issue, an automatic fruit sorting algorithm using deep learning technique was proposed to identify surface defects namely splitting, pitting, and stem-end rot found in mandarin fruits. The fruit sorting algorithm consists of K-Medoids segmentation and Convolutional Neural Network (CNN) classification model. The grayscale images of mandarin fruit surface were captured from an image acquisition system built with Near Infrared (NIR) camera. A preprocessing median filter was applied to remove random noise. After preprocessing, segmentation was carried out using K-Medoids clustering to crop the fruit surface image from the background region. Different CNN models namely VGG-16, InceptionV3 and MobileNet were trained and tested with and without transfer learning approach using the cropped image dataset. After training, the cropped fruit surface image was given to CNN model for defect classification. The classification results of the above models improved significantly after implementing transfer learning method. The VGG-16 model achieved a maximum overall classification accuracy of 90% without transfer learning and 99.53% with transfer learning approach when compared with the InceptionV3 and MobileNet. Overall accuracy of MobileNet improved from 57% to 98% after transfer learning and also it takes minimum time for inference. Considering both the overall accuracy and inference time parameters with the transfer learning approach, the MobileNet is found to be the best model for mandarin fruit sorting application.

Purpose Diabetic retinopathy (DR) is one of the dangerous complications of diabetes. Its grade level must be tracked to manage its progress and to start the appropriate decision for treatment in time. Effective automated methods for the detection of DR and the classification of its severity stage are necessary to reduce the burden on ophthalmologists and diagnostic contradictions among manual readers. Design/methodology/approach In this research, convolutional neural network (CNN) was used based on colored retinal fundus images for the detection of DR and classification of its stages. CNN can recognize sophisticated features on the retina and provides an automatic diagnosis. The pre-trained VGG-16 CNN model was applied using a transfer learning (TL) approach to utilize the already learned parameters in the detection. Findings By conducting different experiments set up with different severity groupings, the achieved results are promising. The best-achieved accuracies for 2-class, 3-class, 4-class and 5-class classifications are 86.5, 80.5, 63.5 and 73.7, respectively. Originality/value In this research, VGG-16 was used to detect and classify DR stages using the TL approach. Different combinations of classes were used in the classification of DR severity stages to illustrate the ability of the model to differentiate between the classes and verify the effect of these changes on the performance of the model.

Crop type mapping by using transfer learning

EEG based multi-class seizure type classification using convolutional neural network and transfer learning

Today breast cancer is the leading type of cancer among women undergoing cancer screening. A slight delay in detecting and diagnosing this disease may result in irreversible convolutions. Histopathological images from the biopsy examination present a large amount of structural information that can signi cantly improve the prognosis for breast cancer. The pathological analysis, which involves the microscopic examination of the histopathological slides, is a challenging task. An automated computer-aided detection (CAD) procedure is inevitable, as it may decrease the pathologist examination time and help detect the disease at an early stage. Lately, deep learning methods using arti cial neural networks are consistently in use to improve the performance of CAD methods. A common practice among recent studies is to use the transfer learning approach of training deep neural network architectures. Transfer Learning is an established learning approach that facilitates a deep neural network to train quickly on a speci c dataset and resolve an interdisciplinary problem. Deep Learning methods employing the transfer learning approach have provided highly competitive results on the datasets consisting of the whole slide images, which are captured generally at high resolutions. However, the performance is not remarkably appreciable on the small and low-resolution image datasets, in particular the datasets that include patch samples. In this direction, the present study proposes a novel domain-speci c learning strategy, Breast Histo-Fusion, which aims to detect breast cancer even from images of low resolution and small size. Further, four state-of-the-art deep CNNs (AlexNet, VGG19, ResNet, and DenseNet) are trained using both learning approaches: Transfer Learning and Histo-Fusion on the IDC dataset. The proposed Histo-Fusion learning approach has improved the discriminating abilities and performance of each deep CNN by providing better results of (AlexNet -95.75%, VGG19-95.96%, ResNet34-96.17%, ResNet50-96.67%, and DenseNet121-97.49%) compared to (AlexNet -90.41, VGG19-90.51%, ResNet34-90.83%, ResNet50-92.27%, and DenseNet121-93.05%) using the transfer learning strategy. As a result, the procedure can help expert pathologists to perform accurate diagnoses and reduce false-positive rates.

Due to the high availability of large-scale annotated image datasets, paramount progress has been made in deep convolutional neural networks (CNNs) for image classification tasks. CNNs enable learning highly representative and hierarchical local image features directly from data. However, the availability of annotated data, especially in the medical imaging domain, remains the biggest challenge in the field. Transfer learning can provide a promising and effective solution by transferring knowledge from generic image recognition tasks to the medical image classification. However, due to irregularities in the dataset distribution, transfer learning usually fails to provide a robust solution. Class decomposition facilitates easier to learn class boundaries of a dataset, and consequently can deal with any irregularities in the data distribution. Motivated by this challenging problem, the paper presents Decompose, Transfer, and Compose (DeTraC) approach, a novel CNN architecture based on class decomposition to improve the performance of medical image classification using transfer learning and class decomposition approach. DeTraC enables learning at the subclass level that can be more separable with a prospect to faster convergence.We validated our proposed approach with three different cohorts of chest X-ray images, histological images of human colorectal cancer, and digital mammograms. We compared DeTraC with the state-of-the-art CNN models to demonstrate its high performance in terms of accuracy, sensitivity, and specificity.

The increasing demands for the reliability of modern industrial equipment and structures necessitate advanced techniques for design, monitoring, and analysis. This review article presents the latest research advancements in the application of machine learning techniques to vibration and acoustic signal analysis from 2015 to 2024. A total of 96 peer-reviewed scientific publications were examined, selected using a systematic Scopus-based search. The main research areas include processes such as modeling and design, health management, condition monitoring, non-destructive testing, damage detection, and diagnostics. In the context of these processes, a review of machine learning techniques was conducted, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory (LSTM), autoencoders, support vector machines (SVMs), decision trees (DTs), nearest neighbor search (NNS), K-means clustering, and random forests. These techniques were applied across a wide range of engineering domains, including civil infrastructure, transportation systems, energy installations, and rotating machinery. Additionally, this article analyzes contributions from different countries, highlighting temporal and methodological trends in this field. The findings indicate a clear shift towards deep learning-based methods and multisensor data fusion, accompanied by increasing use of automatic feature extraction and interest in transfer learning, few-shot learning, and unsupervised approaches. This review aims to provide a comprehensive understanding of the current state and future directions of machine learning applications in vibration and acoustics, outlining the field’s evolution and identifying its key research challenges and innovation trajectories.

1 - Classification of ocular diseases using transfer learning approaches and glaucoma severity grading

Purely, data-driven large scale image classification has been achieved using various feature descriptors like SIFT, HOG etc. Major milestone in this regards is Convolutional Neural Networks (CNN) based methods which learn optimal feature descriptors as filters. Little attention has been given to the use of domain knowledge. Ontology plays an important role in learning to categorize images into abstract classes where there may not be a clear visual connect between category and image, for example identifying image mood — happy, sad and neutral. Our algorithm combines CNN and ontology priors to infer abstract patterns in Indian Monument Images. We use a transfer learning based approach in which, knowledge of domain is transferred to CNN while training (top down transfer) and inference is made using CNN prediction and ontology tree/priors (bottom up transfer ). We classify images to categories like Tomb, Fort and Mosque. We demonstrate that our method improves remarkably over logistic classifier and other transfer learning approach. We conclude with a remark on possible applications of the model and note about scaling this to bigger ontology.

In this work, we evaluated the ability of a CNN-based solution, using transfer learning, to produce an expert-like judgment in recognizing LBBB false defects. We collected retrospectively, MPI polar maps, of patients having small to large fixed anteroseptal perfusion defect. Images were divided into two groups. The LBBB group included patients where this defect was judged as false defect by two experts. The LAD group included patients where this defect was judged as a true defect by two experts. We used a transfer learning approach on a CNN (ResNet50V2) to classify the images into two groups. After 60 iterations, the reached accuracy plateau was 0.98, and the loss was 0.19 (the validation accuracy and loss were 0.91 and 0.25, respectively). A first test set of 23 images was used (11 LBBB, and 12 LAD). The empiric ROC (Receiver operating characteristic) Area was estimated at 0.98. A second test set (18x2 images) was collected after the final results. The ROC area was estimated again at 0.98. Artificial intelligence, using CNN and transfer learning, could reproduce an expert-like judgment in differentiating between LBBB false defects, and LAD real defects.

Recent advances in land use and land cover (LULC) classification with remote sensing imagery are driven by state-of-the-art models such as Convolutional Neural Networks (CNNs). Advanced CNN architecture like ResNet can enhance overall classification performance by incorporating residual skip connections. The integration of 3D feature extraction and ResNet architecture suggests a potential improvement in classification tasks. This paper explores the potential of the 3D ResNet model for LULC classification, comparing it with baseline approaches (Support Vector Machine, Random Forest, XGBoost, 1D CNN, 3D CNN) and state-of-the-art 3D models (3D VGG, 3D DenseNet) using WorldView-2 satellite imagery. The 3D ResNet-18 model, fine-tuned via transfer learning on multispectral images, demonstrates significant improvements in classification performance over machine learning models. It achieves the highest Overall Accuracy (OA) of 99.66% and Kappa Accuracy (KA) of 99.39% on the primary dataset. Despite having slightly lower performance on the external validation dataset (OA:82.89%, KA:80.05%) than 3D DenseNet, it is highly efficient with processing times of 490.2 minutes and 3.6 minutes for both datasets respectively. McNemar’s test results show 3D ResNet and 3D DenseNet have significant differences in classification performance (p<0.05) against other models consistently for both datasets.

Offline Arabic handwritten word recognition: A transfer learning approach

BackgroundBanana (Musa spp.) is the most popular marketable fruit crop grown all over the world, and a dominant staple food in many developing countries. Worldwide, banana production is affected by numerous diseases and pests. Novel and rapid methods for the timely detection of pests and diseases will allow to surveil and develop control measures with greater efficiency. As deep convolutional neural networks (DCNN) and transfer learning has been successfully applied in various fields, it has freshly moved in the domain of just-in-time crop disease detection. The aim of this research is to develop an AI-based banana disease and pest detection system using a DCNN to support banana farmers.ResultsLarge datasets of expert pre-screened banana disease and pest symptom/damage images were collected from various hotspots in Africa and Southern India. To build a detection model, we retrained three different convolutional neural network (CNN) architectures using a transfer learning approach. A total of six different models were developed from 18 different classes (disease by plant parts) using images collected from different parts of the banana plant. Our studies revealed ResNet50 and InceptionV2 based models performed better compared to MobileNetV1. These architectures represent the state-of-the-art results of banana diseases and pest detection with an accuracy of more than 90% in most of the models tested. These experimental results were comparable with other state-of-the-art models found in the literature. With a future view to run these detection capabilities on a mobile device, we evaluated the performance of SSD (single shot detector) MobileNetV1. Performance and validation metrics were also computed to measure the accuracy of different models in automated disease detection methods.ConclusionOur results showed that the DCNN was a robust and easily deployable strategy for digital banana disease and pest detection. Using a pre-trained disease recognition model, we were able to perform deep transfer learning (DTL) to produce a network that can make accurate predictions. This significant high success rate makes the model a useful early disease and pest detection tool, and this research could be further extended to develop a fully automated mobile app to help millions of banana farmers in developing countries.